CHAPTER 25
WAVE CLIMATE CYCLES AND COASTAL ENGINEERING
PRACTICE
Douglas L. Inman,1 Scott A. Jenkins,1 M. Hany S. Elwany1
Abstract
El Nino-Southern Oscilllation (ENSO) events are major
factors in changing wave climate at coastal sites. These events are
triggered by extreme anomalies in the global seasonal climate; and
in recent years they have become quasi-predictable from certain
antecedent conditions. Judicious application of the global concepts
leading to (ENSO) events will improve our understanding and
prediction of wave climate.
Introduction
Design of coastal structures and planning for beach remedial
systems depend upon statistics determined from the study and
measurement of past events and upon models using those statistics
as primary inputs. The value of the statistic is usually assumed to
depend solely upon the accuracy and duration of the observational
series. This leads to common statistics such as a "design wave"
height based on a recurrence history that defines the storm of the
decade and century, and to shoreline changes based upon simple
statistics such as the mean and standard deviation of beach width.
The limitation of our statistic is usually not the quality or
resolution of the observations it is based upon, but rather that the
statistics do not take into consideration the episodic nature of larger,
1
Center for Coastal Studies, Scripps Institution of Oceanography,
La Jolla, California 92093-0209
314
WAVE CLIMATE CYCLES
315
controlling systems that govern the intermediate to long-term wave
climate and sediment budget. For example, wave climate is an end
product of meteorological climate and subject to the cyclicities and
uncertaintities associated with that subject. Beach width is an end
effect of the position within a littoral cell and is subject to all of the
factors that influence the littoral sediment budget, including source,
transport paths and sinks of sediment, as well as the variability in
wave climate. Also, in the case of beaches, there is an element of
inertia that causes beach erosion to be influenced more by the
successive occurrence of clusters of average storms rather than by
the isolated occurrence of a single extreme storm.
Climatic events such as El Nino-Southern Oscillation (ENSO)
effect wave climate on a world-wide basis by altering the global
atmospheric circulation paths, introducing periods of abrupt chaotic
or episodic change. El Ninos are harbingers of more intense wave
climate along the temperate west coast of the Americas, but result
in milder wave climates for the temperate east coasts of the
Americas (Inman and Masters, 1994; Neelin et al 1994). El Ninos
also modify rainfall, causing droughts in some areas and flooding
and maximum sediment yield in others. Along the southern
California coast decades of relatively stable, mild wave climate are
interrupted by El Nino events characterized by groups or clusters of
intense storms and heavy rainfall. The El Nino of 1982/83 and its
associated cluster storms completely denuded beaches that had been
stable for the preceding 30 years. Thus, it is apparent that the usual
statistical techniques based upon limited record lengths for
predicting beach width and littoral drift rates and directions would
lead to conclusions that are invalid for episodic periods of
significant wave climate change.
Seasonal Climate: An Introduction to ENSO Events
The seasonal variations in the exposure of the hemispheres to
the sun produce interannual changes in the duration of daylight and
the angle of the sun's irradiance. These effects modulate solar
heating, resulting in the interannual variation of the earth's
atmospheric pressure field which in turn introduces seasonal
climatic effects. Interannual variations are enhanced by the higher
316
COASTAL ENGINEERING 1996
convective effects of land and the greater concentration of land
mass relative to water in the temperate latitudes of the northern
hemisphere. In January, a large intense area of high pressure forms
over the Asian continent (Figure la). This occurs because the
convective slopes of Asia are inclined away from the sun with short
periods of daylight that shutdown upslope convection, resulting in
subsidence of air mass and an increase in pressure at ground level.
In July, the convective slopes are inclined towards the sun with
longer periods of daylight that increase solar irradiance and fuel
upslope convection and lower surface pressure. Note that, in
contrast to a rigid-lid system, the free motion of air in the
convective atmosphere results in an inverse relation between
temperature and pressure.
The northeast monsoons of southern Asia maximize in
January, mainly in response to downslope anticyclonic flowing air
caused by subsidence under the large area of Asian high pressure
(Figure la). The cold descending air converges with tropical air to
form a wet monsoon. The northeast monsoon in the Indian Ocean
is less intense because of lower pressure gradients and the blocking
effect of the Asian and African continental mountains. During July,
the combination of upslope flow towards the low pressure area over
the Himalaya mountains and the cyclonic geostrophic flow generates
a strong southwest monsoon over the Indian Ocean (Figure lb).
This brings heavy rainfall over India and the high Ethiopian plateau
(Quinn, 1992). The latter causes the Nile River floods that
maximize during July. In contrast to the Indian Ocean, the
southwestern monsoon is weak and relatively dry over southeast
Asia because of lower pressure gradients and blocking by high
inland topography.
The ENSO System
Upon occasion the typical seasonal weather cycles discussed
above are abruptly and severely modified on a global scale. These
intense global modifications are signalled by anomalies in the
pressure fields between the tropical eastern Pacific and Malaysia
known as the El Nino/Southern Oscillation (ENSO) (e.g., Diaz and
Markgraf, eds., (1992). The intensity of the oscillation is often
WAVE CLIMATE CYCLES
317
measured in terms of the Southern Oscillation Index (SOI), defined
as the monthly mean sea level pressure anomaly in mb normalized
by the standard deviation of the monthly means for the period 19511980 at Tahiti, minus that at Darwin, Australia (Figure 2). A
0°
- a. January
90°E
180°
90°W
0°
t
Vm77/////V/A>////77777///////\/////,
c. Pressure Anomaly:
+Ap J SOI Negative (El Nino)
-Ap -i SOI Positive (La Nina)
L -Ap
r +Ap
Figure 1. Seasonal pressure and winds for (a) January and (b)
July. Contours of 1020 mb and 1000 mb are shown around
areas of high (H) and low (L) pressure respectively. Prevailing
winds are indicated by arrows: NEM, SEM, SWM designate
northeast, southeast and southwest monsoons, (c) The pressure anomaly, Ap centered around longitudes 105° E and 105°
W for negative and positive Southern Oscillation Indices (SOI).
320
COASTAL ENGINEERING 1996
Asian low pressure area will be intensified, spinning up a strong
southwestern monsoon in the Indian Ocean, causing increased rain
in India and over the Ethiopian plateau. In July, the air flow over
the northern Indian Ocean is dominated by the large Asian low
pressure system centered over Afghanistan. This intensifies the
southwesterlies and draws the intertropical convergence zone
(ITCZ) north over the Ethiopian plateau. The ITCZ is the zone of
collision between the converging tradewinds of the northern and
southern hemispheres. Its primary features are surface doldrums
and vertical convection with considerable cumulus development,
rain and thunder storms. For intense La Ninas, the resulting
convergence of air masses in the ITCZ brings heavy rainfall that
causes floods on the Nile River (Quinn, 1992). Application of -Ap
anomaly to the southern hemisphere tropics in the vicinity of 105° E
lowers the pressure over northern Australia and New Guinea. This
modifies the pattern of the southeast monsoon and brings heavy
monsoonal rainfall to Australia (e.g., Nicholls, 1992).
Of course, the above suggestions based on an intensified
seasonal pattern are vastly oversimplified and tell us nothing about
why or when ENSO events occur or how long they will last.
However, climatic events characterized by intense El Nifios and La
Ninas affect wave climates on a world-wide basis in the sense that
atmospheric circulation paths are altered globally (Neelin, et. al.,
1994; Philander, 1989). Spectral analysis of coral growth rates
shows that over the past century El Nino events have occurred with
recurrence periods centered approximately on 3 and 7 years with
more intense events occurring every decade or so (Cole et al, 1992;
1993). Although the events are not clearly understood, there is
evidence that ocean-atmosphere interactions act as an oscillator
which drives large-scale ocean waves known as equatorial Kelvin
waves. However, this oscillator interacts with Earth's annual cycles,
producing nonlinear resonances (Jin et al, 1994; Tziperman et al,
1994). The ENSO cycles jump irregularly among these nonlinear
resonances, exhibiting chaotic behavior. Thus, there may be
relatively long periods of uniform annual climate response,
interrupted by periods of abrupt change.
El Nifios are harbingers of more intense wave climates along
the temperate west coast of the Americas, but may result in milder
WAVE CLIMATE CYCLES
317
measured in terms of the Southern Oscillation Index (SOI), defined
as the monthly mean sea level pressure anomaly in mb normalized
by the standard deviation of the monthly means for the period 19511980 at Tahiti, minus that at Darwin, Australia (Figure 2). A
o°
- a. January
180°
90°E
90°W
0°
£v,
C^**
«sW
r
=?^?^^Z^^g?^^izd^^?;
c. Pressure Anomaly:
+Ap J SOI Negative (El Niiio)
-Ap -j SOI Positive (La Nina)
L -Ap
r +Ap
- 80°
W//S///////»/b//w»/»//777PW77&7r7r7TTrX'
0°
90°E
180°
90°W
0°
Figure 1. Seasonal pressure and winds for (a) January and (b)
July. Contours of 1020 mb and 1000 mb are shown around
areas of high (H) and low (L) pressure respectively. Prevailing
winds are indicated by arrows: NEM, SEM, SWM designate
northeast, southeast and southwest monsoons, (c) The pressure anomaly, Ap centered around longitudes 105° E and 105°
W for negative and positive Southern Oscillation Indices (SOI).
3tt
COASTAL ENGINEERING 1996
negative SOI (lower pressure at Tahiti, higher pressure at Darwin) is
known as an El Nino or warm ENSO event, because of the arrival
60° N
60° N
30°
30°
30°
- 30°
60°
60° S
Figure 2. Global distribution of the correlation coefficient of annual
pressure anomalies with simultaneous pressure anomalies at
Jakarta (107° E longitude). Redrawn from Berlage (1957) and
centered on 180° to show the southern oscillation.
of unusually warm surface water off the coast of Peru at the time of
Christmas; hence, the term El Nino. Warm water also occurs along
the coast of California and both regions experience unusually heavy
rainfall. A positive SOI is known as La Nina and it signals the
occurrence of colder than normal surface water in the eastern
Pacific, but stronger southwest monsoons in the Indian Ocean with
heavy rainfall in India and in the Ethiopian plateau. The latter
causes flooding of the Nile River. Thus, the terms warm and wet
vs cold and dry ENSO events that accompany El Ninos and La
Ninas respectively apply to the waters and coastal areas of the
tropical-subtropical eastern Pacific Ocean and not necessarily to
other parts of the globe.
It has been shown that the pressure changes associated with
ENSO events are global in extent, and follow a pattern of alternate
positive and negative correlation coefficients centered around
WAVE CLIMATE CYCLES
321
wave climates for the temperate east coasts of the Americas (Dolan
et al, 1990 and Seymour, 1996). As El Nino events and groups of
intense storms (referred to as "cluster storms") alter the wave
climate, models of the nearshore must be adjusted to these changes
in oder to account for the abrupt and often long-lived coastal
changes induced by these events.
The El Nino system
El Nino systems involve a complex set of interactions
between the atmospheric and ocean circulation at tropical to middle
latitude. The large-scale wind patterns and the companion midlatitude high pressure systems described previously have a
complementary feature in the ocean circulation referred to as
"gyres". The gyres are the dynamic equivalent of the mid-latitude
highs in the atmosphere, and are the result of an elevated sea level
in the center of the ocean basin caused by a convergence of the
Ekman drift from the trades and the mid-latitude westerlies.
A
large pool of warm water is established in the western Pacific which
is ultimately regulated by the strength of the tradewind-driven
component of gyre circulation.
This warm pool extends from the Philippine Islands south to
the mid-Australian coast (Figure 3). Associated with these
prevailing conditions is a stationary area of low atmospheric
pressure, centered on Darwin, Australia, as compared with the
relatively higher sea level pressure in the eastern Pacific from Tahiti
to the coast of the Americas. El Nino events are associated with
the release to the east of the warm water pool and a reversal of the
relative pressure fields at Darwin as compared with Tahiti and the
coast of the Americas (Figure la,c). These changes in the pressure
fields in the lower layers of the atmosphere redirect the upper level
jet streams that steer the tracks of traveling storms and markedly
alter the wave climate of the world's coastline. The sequences
within an El Nino event are illustrated in Figure 3 by numbers (1)
through (5). The trade winds setup and confine a pool of warm
surface water on the western side of the equatorial Pacific Ocean
(1). Fluctuations in the prevailing trade wind intensity may release
a series of eastward-flowing pulses of warm water. The warm El
322
COASTAL ENGINEERING 1996
Figure 3. Schematic illustration of the five sequences leading to a
fully developed El Nino event. See text for description.
Nino event begins when the trade winds relax and release a large
"slosh" from this warm water pool which travels to the east along
the equator (2). These sloshes travel as soliton-like internal features
that are low-mode, baroclinic Kelvin waves channeled in the
equatorial wave guide by the north-south gradients in planetary
vorticity. Planetary vorticity is the spin imparted to a fluid particle
by the local surface-normal vector component of the earth's angular
velocity. These Kelvin waves have a phase speed determined by
the density difference, Ap, at the thermocline, such that their phase
speed is order:
C = [(Ap/p^ghJ m = 150 cm/sec,
where hx is the depth of the thermocline. The equatorial Kelvin
waves pump warm water into the eastern equatorial Pacific (3),
WAVE CLIMATE CYCLES
323
where it is spread poleward by barotropic Kelvin waves with
topographically trapped modes propagating along the continental
margins (4). These waves oppose the flow of the California and
Peru currents. In turn, the shelf-trapped Kelvin waves appear to
excite Rossby planetary waves that propagate slowly to the west
against the flows of the North Pacific Current and the Antarctic
Circumpolar Current (5). In addition to slowing the ocean current
system, the Rossby waves induce sea surface temperature anomalies
which modify surface pressures in the lower atmosphere that
ultimately decrease the intensity of the westerlies, further slowing
the overall circulation of water in the large Pacific gyres.
Sequences (1) through (5) in the El Nino event are all
interactive among themselves and with the atmospheric circulation
in very complex ways that are poorly understood. The conditions
triggering the beginning of an El Nino event or those leading to its
demise are not yet known. However, it is clear that, once started,
the Kelvin-Kelvin-Rossby sequences all provide positive feedback
that enhance each other and work toward a spin-down in the
intensity of the prevailing anticyclonic oceanic gyres.
Changing Wave Climate
ENSO events leading to significant changes in seasonal trends
of wave climate are usually associated with SOI values greater than
about + 0.5. Figure 4 shows hemispheric contour plots of the
pressure height anomalies (meters) for SOI positive (La Nina) and
SOI negative (El Nino). Pressure gradients cause winds, and height
anomalies show where persistent pressure gradients occur. The
strongest pressure gradients occur along the boundaries of the
height anomalies. The height anomalies in the figure show the
mean location and intensities of the pressure changes during the
ENSO events, and their boundaries indicate the likely location of
storm paths. Figure 4a (La Nina) shows large areas of high pressure
over the North Pacific and eastern North Atlantic. These high
pressure areas would enhance storm tracks as shown by the two
arrows for waves apporaching North America and Europe. The El
Nino condition is characterized in Figure 4b by a large area of
negative height anomaly over the North Pacific ocean and a
324
COASTAL ENGINEERING 1996
relaxation in intensity but an increase in area of the positive height
anomaly in the eastern North Atlantic. Thus, the La Nina/El Nino
pattern over the North Pacific is bipolar and would be expected to
give distinct changes in the storm track locations for the two
conditions. In contrast either extreme would tend to enhance storm
tracks approaching the coasts of Europe and the southeastern
Mediterranean coast.
Figure 4. Height anomalies of the 700 mb atmospheric pressure
surface derived from averaging winters (October-March)
between 1947 and 1992 with preceding June-November for 11
SOI of +0.5 or greater (a) and for 12 SOI of -0.5 or less (b)
(data from Redmond and Cay an, 1994). Shading and hatching
indicate areas of negative (L) and positive (H) anomalies with
maximum departure in meters. Arrows show position of
storm-track enhancement associated with the anomalies.
Along many coasts, there are decades of relatively stable
weather interrupted by shorter periods characterized by more
variable weather, often accompanied by severe storms. The most
recent period of mild-stable weather along the southern California
325
WAVE CLIMATE CYCLES
coast occurred during the 30 years between the mid-1940's and mid1970's when mild La Nina type weather prevailed. Winters were
moderate with low rainfall (Figure 5), and winds were predominantly from the west-northwest. The principal wave energy was
from Aleutian lows having storm tracks which usually did not
reach southern California (Figure 4a). Summers were mild and dry
with principal wave energy coming from the southern hemisphere.
5?
s
<0
30 1
f— Dry
x: 20 .
u
c
1
1
1
> fr
> I
> tT
Wet' Dry
1
Wet
> jT
1
Dry > t Wet'
- 40
- 0
£
o
-40
-80
2000
Figure 5. Cumulative residual rainfall in San Diego illustrating wet
and dry periods that correlate with El Nino trends.
The wave climate in Southern California changed, beginning with
the El Nino years of 1979/80 and 1982/83. The prevailing
northwesterly winter waves have been replaced by waves
approaching from the west (Figure 4b), and the previous southern
hemisphere swell waves of summer have been replaced by tropical
storm waves from the waters off Central America. The net result
appears to be a decrease in the southerly component of the
longshore transport of sand that prevailed during the preceding
thirty years (Inman and Masters, 1994). Previous southward drift
326
COASTAL ENGINEERING 1996
rates of 200,000 m3/yr. have decreased to 50,000 m3/yr. and have
reversed direction for protracted periods of time.
Conclusions
The clue to successful application of time series to engineering practice lies in long record lengths, and in the identification of
supposedly anomalous trends. These anomalous trends can be
identified by observables that either have intrinsically long record
length or correlate with certain cause and effect mechanisms. For
example, rainfall records, tree rings, railroad surveys and old newspaper articles can be used to correlate beach width and wave
erosion with past El Nino's. Fluctuations in rainfall and wave
intensity both follow the ENSO cycle which links them. For the
modern records, a number of techniques are available to help
elucidate this problem of cause and effect, including spectral and
cross-spectral approaches and numerical progressions such as those
resulting in cumulative residuals ( Figure 5).
Acknowledgments
This research was funded by the Office of Naval Research,
Ocean Modeling and Prediction program, under grant N00014-95-1 0005 with the University of California, San Diego.
References
Berlage, H. P., 1957, "Fluctuations of the general atmospheric
circulation of more than one year, their nature and
prognostic value," Koninklijk Nederlands Meteorlogisch Institut,
Mededelingen en Verhandelingen, 69, 152 p.
Cole, J. E., G. T. Shen, K. G. Fairbanks and M. Moore, 1992,
"Coral monitors of El Nino/Southern oscillation dynamics
across the equatorial Pacific," pp. 349-375 in H. F. Diaz and V.
Markgraf, eds., El Nino, Historical and Paleoclimatic Aspects of
the Southern Oscillation, Cambridge Univ. Press, Cambridge,
U. K., 476 p.
WAVE CLIMATE CYCLES
327
Cole, J. E., R. G. Fairbanks and G. T. Shen, 1993, "Recent
variability in Southern oscillation: isotopic results from a
Tarawa Atoll coral," Science, v. 260, p. 1790-93.
Diaz, H. F. and V. Markgraf, eds., 1992, "El Nino, historical and
Paleoclimatic aspects of the Southern oscillation," Cambridge
Univ. Press, Cambridge, U. K., 476 pp.
Dolan, R., D. L. Inman and B. Hayden, 1990, "The Atlantic coast
storm of March 1989," Jour. Coast. Res., v. 6, n. 3, p. 721-25.
Inman, D. L. and P. M. Masters, 1994, "Status of research on the
nearshore," Shore and Beach, v. 72, n. 3, p. 11-20.
Jin, F-F, J. D. Neelin, and M. Ghil, 1994, "El Nino on the Devil's
staircase: annual subharmonic steps to chaos," Science, v.
264, p. 70-72.
Neelin, J. D., M. Latif and F-F Jin, 1994, "Dynamics of coupled
ocean-atmosphere models: the tropical problem," Annual Rev.
Fluid Mech., v. 26, p. 617-659.
Nicholls, N., 1992, "Historical El Nino/Southern oscillation
variability in the Australasian region," pp. 151-173 in H. F.
Diaz and V. Markgraf, eds., El Nino, Historical and
Paleoclimatic Aspects of the Southern Oscillation, Cambridge
Univ. Press, Cambridge, U.S., 476 pp.
Philander, S. G., 1989, "El Nino, La Nina, and the southern Oscillation," Int. Geophy. Ser., v. 46, Acad. Press, S.D., 293 pp.
Quinn, W. H., 1992, A study of Southern Oscillation-related
climatic activity for A. D. 622-1900 incorporating Nile River
flood data," p. 119-194 in H. F. Diaz and V. Markgraf, eds, El
Nino, Historical and Paleoclimatic Aspects of the Southern
Oscillation, Cambridge Univ. Press, Cambridge, UK, 476 pp.
Redmond, K. T. and D. R. Cayan, 1994, "El Nino/Southern
oscillation and western climate variability," 6th AMS
Conf. on Climate Variations, Nashville, Tenn.
Seymour, R., 1996, "Wave climate variability in southern
California," Jour. Waterway, Port, Coastal, and Ocean
Engineering, v. 122, n. 4.
Tziperman, E., L. Stone, M. A. Crane, and H. Jarosh, 1994, "El
Nino chaos: overlapping of resonances between the seasonal
cycle and the Pacific Ocean-atmosphere oscillator," Science,
v. 264, p. 72-74.